The activity of the D vitamins (vitamins D.sub.3 or D.sub.2) in regulating calcium, metabolism and normal bone growth and development is known to require metabolism, of parent vitamin to certain hydroxylated forms. Specifically, it has been established that 1.alpha.,25-dihydoxyvitamin D.sub.3 (1,25-(OH).sub.2 D.sub.3), the dihydroxylated metabolite normally formed from vitamin D.sub.3 in the animal or human, is the active species responsible for stimulating calcium transport in the intestine, and calcium resorption from bone (bone mobilization), thereby regulating the overall blood calcium level of the organism. (These calcium-related activities of vitamin D metabolites or analogs will, in the following description, be referred to collectively as the `calcemic activity` or `calcemic action` of the compounds). Certain structural analogs of 1,25-(OH).sub.2 D.sub.3, such as for example, 1.alpha.-hydroxyvitamin D.sub.3, 1.alpha.-hydroxyvitamin D.sub.2, 1.alpha.,25-dihydroxyvitamin D.sub.2, or fluoro-substituted derivatives of 1,25-(OH).sub.2 D.sub.3, are also known as highly active calcemic agents, and as a result 1,25-(OH).sub.2 D.sub.3 and its active analogs have been used, or proposed, as pharmaceuticals in the propylaxis or treatment of various calcium metabolism and bone disorders, such as renal osteodystrophy, vitamin D-resistant rickets, or osteoporosis and related diseases.
More recently, it has been discovered that 1,25-(OH).sub.2 D.sub.3, in addition to its well-known `calcemic action` discussed above, also expresses other biological functions. For example, it has been found that 1,25-(OH).sub.2 D.sub.3 and closely related analogs (1.alpha.-OH-D.sub.3, 1,25-(OH).sub.2 D.sub.2, fluoro-substituted analogs, etc.) are capable of inducing cellular differentiation [Abe et al., Proc. Natl. Acad. Sci. USA 78, 4990 (1981); Honma et al., Proc. Natl. Acad. Sci USA 80, 201 (1983)]. Specifically, 1,25-(OH).sub.2 D.sub.3 and its analogs has been shown to inhibit the proliferation of malignant cells grown in culture (e.g. human leukemia cells) and induce their differentiation to normal macrophage-type cells. (These types of activities will henceforth be referred to collectively as the "differentiation activity" of vitamin D compounds.) Because of their remarkable potency as differentiation-inducing agents, these vitamin D derivatives are potentially useful for anticancer agents, and their use for the treatment of human leukemias has indeed been proposed (Suda et al., U.S. Pat. No. 4,391,802). However, even though these compounds are highly effective in differentiating malignant cells in culture, their equally high calcemic action in vivo limits or precludes their use as practical anticancer agents. Thus, 1,25-(OH).sub.2 D.sub.3 or its fluorinated derivatives are exceedingly potent cell differentiation agents, but they also are the most potent compounds with respect to calcemic activity, and at the levels required in vivo for effective use as anticancer (e.g. antileukemic) agents, these same compounds can produce dangerously elevated blood calcium levels by virtue of their inherent calcemic activity. Other known vitamin D derivatives show a similar correspondence between differentiation activity and calcemic activity, and their practical use as potential anticancer agents, therefore, is subject to the same limitations and hazard.
These observations clearly indicated a need, and have stimulated a search, for compounds with greater specificity and selectivity of action as anticancer agents, i.e. for compounds with an improved differentiation/calcemic activity ratio, and recent work has, indeed, led to the preparation of several vitamin D analogs with enhanced differentiation activity. It has been found for example, that certain 1,25-(OH).sub.2 D.sub.3 homologs, in which the side chain is extended by one carbon (either within the chain or at its terminus) exhibit a markedly higher differentiation activity (about 10 times) for leukemia cells in culture than 1,25-(OH).sub.2 D.sub.3 itself [DeLuca et al., U.S. Pat. No. 4,717,721; Ostrem and DeLuca, Steroids 49, 73-102 (1988); Ostrem et al., J. Biol. Chem. 262, 14864 (1987)]. However, these homologs are still extremely potent calcemic agents, exhibiting calcemic activities approximately equal to that of 1,25-(OH).sub.2 D.sub.3. These compounds, therefore, are characterized by an improved differentiation/calcemic activity ratio, but they do not overcome the problem of the undesired potent calcemic action discussed above. Other vitamin D-related compounds, reported to have preferential differentiation activity, have been prepared [see Ostrem et al., supra; Kubodera et al. Chem. Pharm. Bull. 34, 2286-89 (10,86); Ikekawa et al. Chem. Pharm. Bull 35, 4362 (1987)], but these are structurally distinct and different from the compounds of the present invention.